Unverified50% confidenceFactExact time
Holm correction sorts p-values from smallest to largest and compares them in turn against the adjusted significance threshold α/(m-i+1), stopping subsequent tests as soon as one test is not significant; it is less conservative and has higher statistical power than Bonferroni correction, and does not require the tests to be independent.
1
Sources
50%
Confidence
Long-term
Relevance
11/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn A/B testing, accuracy differences of less than 2% are usually hard to distinguish, and a larger sample size is needed to reach reliable conclusions.75% similarUnverifiedAlthough sample splitting guarantees the theoretical properties, it reduces the sample size used to compute the test statistic, which may weaken test power in small-sample settings.72% similarUnverifiedBonferroni 校正是最保守的多重比较校正方案，将显著性阈值除以比较次数；Benjamini-Hochberg FDR 校正更宽松，控制所有显著结果中假阳性的比例71% similarUnverifiedIf the highest score after reranking is below a threshold validated against a test set, the model should be told that no relevant content was found.71% similarUnverifiedThe benchmark improvement from 7.9% to 48.1% refers to the change in the model's accuracy on a certain math benchmark; 7.9% is close to random guessing, while 48.1% indicates the model has substantive problem-solving ability.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1137685API
curl https://kongchang.com/api/v1/knowledge/claims/1137685MCP
get_claim(id=1137685)